Performance assessment of college students higher vocational mathematics education using fuzzy evaluation model

Abstract

Higher vocational mathematics education is advanced and related to real-time applications providing vast knowledge. Teaching and training peculiar mathematical problems improve their educational and career-focused performance. Therefore optimal performance assessment methods are required for reducing the lack of knowledge in mathematics learning. This article hence introduces an Articulated Performance Assessment Model (APAM) for consenting mathematics assessment. In this model, fuzzy optimization is used for consenting different factors such as understandability, problem-solving, and replication. The understandability is identified using similar problem progression by the students, whereas replication is the application of problem-solving skills for articulated mathematical models. From perspectives, problem-solving and solution extraction is the theme that has to be met by the student. The assessments hence generate a perplexed outcome due to which the fuzzy optimization for high and low-level understandability is evaluated. The optimization recommends the change in varying steps in problem explanation and iterated replication for leveraging the students’ performance. This process swings between irrelevant and crisp inputs during fuzzification. In this process, the crisp inputs are the maximum replications produced by the students for better understanding. Therefore, the proposed model is evaluated using efficiency, maximum replication, fuzzification rate, and analytical time.

Keywords

Fuzzy optimization mathematics education performance evaluation student assessment

1 Introduction

Mathematics education requires various skills and knowledge to learn certain topics and problems. College student learns mathematics as a subject that requires proper skills and potential among the students. Mathematics effectively builds mental discipline and logical reasoning capability for college students [1]. Analyzing mathematics education is an important task to perform in every educational institution. Programming-based techniques are mostly used for analyzing processes [2]. The analyzing technique first identifies the important key sets and values of skills that are available among the students. Certain questionnaires and tests are conducted for the students, which provide optimal information about mathematical skills and knowledge [3]. The programming-based technique increases the accuracy and efficiency in analyzing, producing important data to improve students’ performance and development range [4]. Realistic mathematic education (RME) is also analyzed among the students. RME provides feasible instruction and knowledge to students for solving real-time problems of students. RME enhances the effective range of decision-making skills of college students. REM improves college students’ performance and knowledge level [5, 6].

Performance assessment is a process that identifies the skills and knowledge of people through various performance tasks. Performance assessment provides relevant information for performance developing systems. College students also use performance assessment for mathematics subjects [7]. Performance assessment uses analysis techniques that analyze college students’ exact mathematical knowledge and skill sets. Monthly tests, practices, and other kinds of tests are conducted for the students [8]. Tests provide feasible information, which is classified based on scores and performance. Mathematics subjects contain certain key values required for every college student [9]. The key values required proper skills and knowledge to solve sums and problems in examinations. Student response types are also analyzed, which produces the necessary data for the performance assessment process [10]. The performance assessment systems are mainly used for the development and management processes. Both pre and post-test scores are evaluated for the assessment process, which reduces the complexity of computation and development systems. The evaluated data contains the exact information on the mathematical skills of college students [11].

Fuzzy models are used for the performance assessment process. A fuzzy logic model is mostly used to improve the accuracy of decision-making systems. The fuzzy logic model identifies the exact content for the mathematics performance assessment process [12]. A heuristic technique is implemented in the fuzzy logic model that analyzes the data which are presented in the database. The heuristic technique provides feasible performance assessment data, reducing the evaluation process’s latency [13]. The fuzzy logic model detects the complex data which are required to solve certain problems in the decision-making process [14]. The fuzzy logic model reduces the overall time and overhead in the computation and decision-making process. An analytical method is also used in the fuzzy model, which identifies the effective datasets for the performance assessment process [15]. The analytical method-based fuzzy model improves the efficiency and performance range of the assessment process. The statistical technique is also implemented in the fuzzy logic model to understand the exact data content required for further processes. The statistical technique uses various tools to evaluate the actual performance level of students in mathematics subjects [13, 16]. Henceforth by assimilating the fuzzy concept, this article introduces an articulated performance assessment model for validating the teaching model efficiency. This is particular about mathematics and the contributions are listed below:

Proposing a teaching-focused assessment model for different mathematical models suitable for different grades.

Applying the fuzzy paradigm for optimizing the model failures and improving the recommendation for leveraging its performance.

Performing a wide assessment study using an appropriate dataset for validating the proposed methods’ efficiency.

2 Related work and motivation

Im and Jitendra [17] introduced proportional reasoning for analysis using mathematical learning disabilities (MLD). The introduced method is a schema-based instruction method that identifies the proportions and capabilities of students. MLD identifies students’ written explanation, producing relevant data for the development process. Error analysis is also used here to detect similar data relevant to mathematics. The introduced analysis method improves the efficiency and performance range of the system.

Legesse et al. [18] proposed a discourse-based instruction method for mathematics statistics. The main aim of the proposed method is to understand the procedure and conceptual probabilities of mathematics. Important key values and variables are collected using a researcher-constructed instrument. Both time and complexity in the identification process are reduced, enhancing the systems’ feasibility. Experimental results show that the proposed method increases the understanding capabilities of students.

Huey [19] designed a standard-based grading policy for secondary mathematics. The proposed policy identifies the performance and grading scores of students. The standard-based policy provides various practices to train students, reducing the understanding process’s complexity. The designed policy analyzes the behaviors and activities of students. The proposed policy provides necessary data for the performance improvement process, which increases the knowledge level of students.

Song et al. [20] proposed a prediction method for mathematics and arithmetic performance. The main goal of the proposed method is to identify the exact longitudinal relationships between young students. A cross-lagged analysis is implemented in the proposed method that analyzes students’ actual skills. The proposed method also predicted the arithmetic fluency and knowledge ratio of students. The proposed method achieves high accuracy in the prediction that improves students’ experience and performance range.

Hidayat et al. [21] developed a meta-cognitive behavior and mathematical model. The developed model is mostly used to identify the impacts of performance goals. Confirmatory factor analysis is used here to analyze the database’s factors and features. The analysis method identifies students’ program status, producing feasible data for the behavior detection process. When compared with other models, the proposed model increases the mathematical competency among the students.

Wang et al. [22] designed a self-determination theory-based motivation prediction method for college students. The main aim of the proposed method is to increase the motivational aspects among the students. College mathematical achievements and competition scores are evaluated that provide appropriate information for the motivational process. Experimental results show that the proposed provides various motivational practices to students, improving their mathematics performance.

Moreno and Pineda [23] introduced an automated formative assessment framework for mathematical courses. Learning analytics is also used here to analyze the relevant data presented in a database. Assessment item prediction is a complicated task to perform in management systems. Mathematics courses contain various variables and features which are managed in a database. The introduced framework achieves high accuracy in the assessment, providing feasible data for further processes.

Nguyen-Huy et al. [24] proposed a student performance prediction method for advanced engineering mathematics courses. A multivariate copula model is used in the prediction method that examines the scores of students. The copula model constructs the method based on certain functions and conditions. The proposed method increases the accuracy of the decision-making process. The proposed method reduces the complexity of the computation and identification process. Compared with other methods, the proposed method maximizes the performance prediction accuracy.

Wijaya et al. [25] developed a meta-analysis for student mathematics achievements. The main goal of the developed method is to analyze the datasets which are required for various processes. A meta-analysis system also analyzes E-books effects. A quantitative approach is implemented in the analysis of achievements that are collected from the management systems. The developed method improves students’ performance and effectiveness range in mathematics subjects.

Ding and Homer [26] designed a multilevel model in the program for international student assessment (PISA). The proposed model is mainly used to analyze the mathematical performance level of students. The multilevel model identifies the reading performance range of students, which provides feasible data for further prediction and development processes. Score interpretation is also evaluated, reducing the computation process’s latency. The designed model increases the understanding capabilities and knowledge ratio of students.

Sun et al. [27] proposed a digital game-based machine-learning model for mathematics. The proposed model is widely used in primary education institutions. Scaffolding strategies are used here that detect the important key values and factors for the teaching process. Various questionnaires are conducted among the students that produce relevant scores and data for the modeling process. The proposed model enhances the performance and efficiency range of mathematical skills among the students.

Lahdenperä et al. [28] developed a mixed-method approach for undergraduate mathematics students. The developed approach aims to understand the mathematical learning environments of students. Self-level and course-level difficulties are analyzed, producing the necessary information for the performance development process. The proposed approach also reduces the difficulties in the learning process. The developed approach reduces the complexity of learning and improves the learning process’s flexibility and efficiency level.

Sabir et al. [29] examined a Susceptible-Infectious-Quarantined (SIQ)-based COVID-19 mathematical model with the help of Artificial Neural Networks and Levenberg-Marquardt backpropagation. This model can mimic the intricate dynamics between susceptible, infected, and quarantined individuals. The results demonstrate a good correlation with the reference dataset, demonstrating accuracy and dependability. Still, more investigation may be needed into questions like sensitivity analysis depth, lockdown intensity exploration, and scenario validation.

Cascella et al. [30] introduced a new analysis approach as an investigation approach for foreign students. The actual goal of the introduced approach is to identify the features and items that favor native students over foreign students. Foreign students face various problems during studying, which reduces their performance level of the students. An item-level analysis is also used here that analyzes the cause and situation of problems which foreign students face. The introduced approach reduces the favoring among students.

Sabir et al. [31] employed a Bayesian regularization neural network (BRNNA) to address the fractional-order Layla and Majnun model (MFLMM). To gauge how well the BRNNA performs, this study employs a stochastic method based on soft computing. The outcomes demonstrate the BRNNA’s efficiency in solving the MFLMM, with reduced absolute errors and consistent and reliable performance. However, the BRNNA’s sensitivity to model parameters and its efficacy on real-world data warrant additional investigation.

The aforementioned methods discussed above foresee probability-based input analysis in [18 , 26] and behavior models in [19 , 27]. The difference is that the probability validates multiple models and the students fitting its teaching style. This readily identifies multiple possibilities for suppressing variations in recommendations. The behavior model takes more complexity due to its adaptable policies and student-dependent modifications. However, the adaptation of different teaching styles requires multivariate analysis and unified assessment of different mathematics teaching models as in [24, 27]. Mathematical models of COVID-19 have been examined using artificial neural networks, although more work is required in this area [29]. While Bayesian regularization neural networks are employed for fractional-order Layla and Majnun models [31], a novel method is needed to understand the difficulties encountered by international students. Therefore, a novel approach, fuzzy optimization for mathematical performance evaluation, covers sensitivity analysis gaps and analyzes efficiency, replication, and analytical time, delivering a comprehensive picture of academic performance dynamics. Therefore by considering this process, the proposed model is finalized for identifying teaching replication based on unified factors. Such a unified process results in defacing autonomous recommendations regardless of the student’s learning type and understandability.

3 Articulated performance assessment model

To provide a large knowledge for college students, mathematics education should be taught, which is advanced and connected to real-time applications. Teaching and training specific mathematical problems can enhance their educational and career-focused performance. Hence, optimal performance assessment methods are required to decrease the lack of knowledge in mathematics learning. So here Articulated Performance Assessment Model (APAM) is introduced for approving different factors such as understandability, problem-solving, and replication. Fuzzy optimization is done based on the observation that teachers make judgments based on indistinct and given information. It represents fuzzy information that can identify, recognize, and interpret. Fuzzy optimization is an optimization problem in AI, fabricating, and management. Fuzzy optimization is regulative and, as a mathematical model, deals with fluctuating ambiguity and information inadequacy uncertainty. Most of the brochure calls these ambiguities equivocal and obscure, respectively. Fuzzification is the fuzzy optimization process of authorizing a system’s numerical input to fuzzy sets with some degree of membership. This degree of membership may be everywhere within the given interval. Fuzzy logic attempts to solve problems with a clear, unspecific span of data and pragmatics, making it possible to attain an order of accurate decisions.

The APAM model is an organized and iterative method that employs several instructional models (MDLs) for grades 3–8. Simulating classroom exercises and testing students’ ability to apply what they’ve learned can gauge how well they grasp mathematical concepts. An integral part of this method is having students repeat their solutions to problems repeatedly to see how well they hold up to repeated testing of their originality. The model uses fuzzy optimization, which involves the transformation of numerical inputs into fuzzy sets to account for issues of interpretability, solvability, and reproducibility. It aids in overcoming assessment-related ambiguity and uncertainty. Through an iterative method, the model guides students to maximize replications while improving comprehension. Inadequate replications or flaws prompt a reconsideration to boost fuzzy optimization and increase student comprehension. The suggested APAM model aims to enhance mathematics performance evaluations by centering on subtle components and using fuzzy optimization for decision-making. It demonstrates significant gains in efficiency, replication, fuzzification rate, and recommendation ratio. The models’ overview is portrayed in Fig. 1.

Fig. 1

Model overview.

Here in this method, the mathematics problems and techniques are educated by the students by the teachers after teaching the performance of the students is evaluated. In this performance, the understandability, problem-solving, and replication are consented to by using the Fuzzy optimization technique. Understandability is the one that is identified by using the same problem progression by the students. Replication is the assessment of problem-solving skills for mathematical models. From this, both the solution evaluation and the students’ problem-solving skills can be assessed. There are things consented to by fuzzy optimization. From this, the understanding level of the students can be calculated, which is high level or low level. The optimization helps in changing the problem explanation and iterated replication for manipulating student performance. This procedure results in irrelevant and crisp input during fuzzification. The crisp input is the one that is the maximum replications delivered by the students for better understanding. If there is no maximum replication by the students, then the modification is done by fuzzy optimization to find the better crisp input by performing the reassessment process. Here the mathematical models are educated by the teachers to the students. After teaching, the performance of the students can be evaluated. The process of evaluating the performance of the students can be explained by the following Equation (1):

$A = (a_{1}, a_{2}, \dots a_{n})$ (1)

Where A is denoted as the calculation of the performance of the students. From the performance evaluation, understandability, problem-solving, and replication are determined. The understandability ability of the students is evaluated by providing a similar problem to solve. From this, how the students understand the problem and how they solve it can be evaluated. The mathematical models and problems that are educated by the teachers they should understand the students for the development of problem-solving skills. In this understandability checking process, the level of the student’s progression will be assessed by giving a similar pattern of the problem, which the teachers teach. The students how they understood the mathematical models and problems and how they used them for solving the problems can be determined in this phase. The skills of the students can be evaluated by checking their understanding of the mathematical models or problems.

To verify their understandability skills, a similar type of problem will be given to the students to solve. They take the time to solve the problems and the way they understand the problems could be determined to evaluate their understandability. The same problem taught by teachers while teaching the mathematical models will be given to the students during their evaluation of their performance. The student’s level of understandability of mathematical models or problems will be evaluated from their performance. The progression of the same type of problems educated to the students used in determining the lucidity level from their performance. The way students think to solve given problems and the time they use to identify the pattern of the problem and solve it can be identified in this understandability evaluation process. A similar way of problems which are taught during their teaching time will be given to the students for solving. The student should recognize the pattern of the problem and solve the problems by using the mathematical models taught by the teachers. The process of evaluating the understandability of the student will be explained by the following Equation (2): $i^{B} = [\begin{matrix} \begin{matrix} {ib}_{11} & {ib}_{12} & \begin{matrix} \dots & {ib}_{1 n} \end{matrix} \end{matrix} \\ \begin{matrix} \begin{matrix} {ib}_{21} \\ ⋮ \end{matrix} & \begin{matrix} {ib}_{22} \\ ⋮ \end{matrix} & \begin{matrix} \begin{matrix} \dots \\ ⋮ \end{matrix} & \begin{matrix} {ib}_{2 n} \\ ⋮ \end{matrix} \end{matrix} \end{matrix} \\ \begin{matrix} {ib}_{n 1} & {ib}_{n 2} & \begin{matrix} \dots & {ib}_{nn} \end{matrix} \end{matrix} \end{matrix}] = i^{{(b_{fl})}_{n \times n}}$ (2)

Where i^B is denoted as the evaluation of the understandability of the students; f is denoted as the mathematical problem, and l is denoted as the models. Now problem-solving skills can be determined by the process. By knowing this, the understandability and the replication can be identified easily. The performance can evaluate the problem-solving skills of the students they perform after teaching the mathematical models and problems. The efficiency of the student’s skills during their problem-solving process can also be determined. From the performance, the skills to solve the problem and the way their understandability can be evaluated during this process. By using this and understandability first part of the evaluation can be done. And by this problem-solving and replication, the second part of the performance evaluation is done. The performance split is illustrated in Fig. 2.

Fig. 2

Performance split for different models.

Their performance can identify the problem-solving technique of the students. After teaching the mathematical problem or models, the same pattern of the problems will be given for the assessment process. From this, the students understand the pattern and do further steps to solve the problems. The students’ time taken to understand the problem and solve the problem will be estimated to identify their problem-solving skills. The students should think of an effective way to solve the given problem within a short time. From this, effective problem-solving skills can be identified and further steps are taken to improve their knowledge if they lag somewhere. This problem-solving skill is the base for both the understandability and the replication evaluation process. By keeping this problem-solving as a base, the two phases of the evaluation process are done (Fig. 2). The process of identifying the problem-solving skills of the students can be explained by the following Equation (3):

$i^{l_{j}} = \frac{1 - i^{x_{1 j}}}{i^{x_{1 j}}}, 0 \leq, l_{ij} \leq 1 for allj = 1, 2, \dots n$ (3)

Where x_1j is denoted as the evaluation of the problem-solving skill. Now the replication can be evaluated from their performance. By using the problem-solving skill of the students, the replication is identified. Replication is the application of problem-solving skills for mathematical models. Replication is the one where the different kinds and patterns of problems will be given to the students to check their skills. The different iterations of the problem will be given to the student and check how they solve it by using the mathematical models or problem. The varying steps in problem explanation and iteration replication are given for the students to estimate their performance. This is the second part of the evaluation process. The student’s efficiency in solving the problems can be evaluated through this replication.

Replication is the process of giving different patterns of problems to the students to check whether they can solve them by thinking innovatively. From this, the efficiency of their problem-solving technique can be evaluated. The way students think to solve the replication is given to check the performance. The students’ performance can be evaluated by extracting the understandability, problem-solving, and replication. These can be determined to evaluate the efficiency of the student’s knowledge after teaching the mathematical models and problems. The process of identifying the replication from the student’s performance can be explained by the following Equations (4) & (5): $i^{L} = [\begin{matrix} \begin{matrix} i l_{11} & i l_{12} & \begin{matrix} \dots & i l_{1 n} \end{matrix} \end{matrix} \\ \begin{matrix} \begin{matrix} i l_{21} \\ ⋮ \end{matrix} & \begin{matrix} i l_{22} \\ ⋮ \end{matrix} & \begin{matrix} \begin{matrix} \dots \\ ⋮ \end{matrix} & \begin{matrix} i l_{2 n} \\ ⋮ \end{matrix} \end{matrix} \end{matrix} \\ \begin{matrix} i l_{n 1} & i l_{n 2} & \begin{matrix} \dots & i l_{n n} \end{matrix} \end{matrix} \end{matrix}] = (i^{l_{f l}}) f o r i = 1, 2, \dots n$ (4) $\begin{matrix} G^{l} = [\frac{a_{i 1}}{a_{il}}, \frac{a_{i 2}}{a_{il}}, \dots . . \frac{a_{in}}{a_{il}}] \\ G^{l_{ij}} = \frac{a_{ij}}{\sum_{j} a_{ij}} \end{matrix}}$ (5)

Where L is denoted as the different pattern of the problem given to the students; G is denoted as the process of evaluating the replication; a_ij is denoted as the varying step in problem explanation, and ∑_j . is denoted as the process of evaluating the iterated application. Now from the evaluation process, the output is given to the fuzzy optimization to determine the level of the student’s understandability. The output of the understandability of the base of problem-solving is given to the fuzzy optimization process. And also, the output of the application from the same base of problem-solving is sent to the optimization process. Fuzzy is used to evaluate the vagueness in the student’s performance. It is also used to establish the efficiency and the level of thinking during the performance. It makes the decision based on the evaluation process. It is more useful as the input given can be weighed in applying the membership functions. The optimization process flow is presented in Fig. 3.

Fig. 3

Optimization process flow.

Fuzzy optimization is used to identify the decision based on the understandability level of the students. From this identification, the students’ high level and the low level of understandability can be determined. The optimization is used to evaluate the performance of the students based on mathematical problems and models. From this optimization technique, the level of the understanding skills of the students and the level of their performance can be evaluated (Fig. 3). The process of fuzzy optimization from the output of the evaluation process is explained by the following Equations (6) & (7):

$\begin{matrix} Z_{1} = \frac{a_{il}}{a_{il} + a_{i 2}} \\ Z_{2} = \frac{a_{il}}{a_{il} + a_{i 3}} \\ Z_{A} = \frac{a_{il}}{a_{il} + a_{in}} \end{matrix}}$ (6) $V_{ij} = \frac{1 - | a_{ij} - a_{i} |}{\sum_{xy} | a_{ij} - a_{i} |}$ (7)

Where Z is denoted as the output of the evaluation process; V_ij is denoted as the process of fuzzy optimization. From the output of the fuzzy optimization process, the high level and the low level of the understandability of the students can be evaluated. From the students’ innovative thinking, the understandability level will be high in the performance. When the students have a high level of understandability, they can innovatively solve problems and improve their performance output. The perfect way of solving the problem makes a high understandability level. The perfection of problem-solving in a short time makes their understandability level high. The output of their performance through fuzzy optimization helps in identifying the level of understandability. The assessment hence provides the outcome due to the fuzzy optimization for the high and low levels of understandability evaluated. From the output of the fuzzy optimization, the level of understandability can be determined through the student’s performance. From the evaluation process, the optimization technique is used to identify the problem-solving skills of the students and the number of replications they are doing perfectly. The process of evaluating the high-level understandability is explained by the following Equation (8):

$\begin{matrix} V_{m} = [\begin{matrix} 1^{v_{11}} & 1^{v_{12}} & \dots & 1^{v_{1 n}} \\ 1^{v_{21}} & 1^{v_{22}} & \dots & 1^{v_{2 n}} \\ \dots & \dots & \dots & \dots \\ 1^{v_{m 1} 1} & 1^{v_{m 1} 2} & \dots & 1^{v_{m 1} n} \\ 2^{v_{11}} & 2^{v_{12}} & \dots & 2^{v_{1 n}} \\ \dots & \dots & \dots & \dots \\ 2^{v_{m 1} 1} & 2^{v_{m 2} 2} & \dots & 2^{v_{m 2} n} \end{matrix}] \\ = {(v_{i j})}_{m \times n} f o r a l l i = 1, 2, \dots, n \end{matrix}$ (8)

Where V_m is the process of evaluating the high level of understandability. Now a low level of understandability is found in the performance output. If the student lacks knowledge while solving the problem, it tends to be less understandable. If the students did not understand the contents of the mathematical problems and models, it will be visibly seen in their performance. Through fuzzy optimization, the level of the student’s understanding of the mathematical problems will be determined. These outputs can be evaluated to find the crisp inputs to attain the maximum replications with the efficacious understandability level of the students. During the evaluation process, the performance will have some time lags and innovative ways of thinking. The students may get less understanding during the teaching process of mathematical models and problems. The process of attaining a low level of understandability in the students is explained by the following Equation (9): ${\begin{matrix} x_{j} = \sum_{i} x_{ij} \\ y_{i} = \sum_{i} x_{ij} \end{matrix}$ (9)

Where x_j is denoted as the evaluation of low-level understandability; y_i is denoted as the estimation of the understandability during problem-solving. From this output, the crisp input can be determined. During the fuzzification process, the crisp input can be evaluated from the output of the understandability. The process of replication in the performance can increase the way of crisp input. The students are made to perform the replication process better, which is given by the iterated progression of the problem. The way of solving the problem can be improved; thus, crisp inputs are provided. The crisp inputs are the ones that are obtained from the fuzzy optimization output. The output of the level of understandability evaluation process decides the crisp input by the maximum replication process of the students’ better understanding skills. The process of providing the crisp input is explained by the following Equations (10) & (11): $D = [\begin{matrix} \begin{matrix} d_{11} \\ d_{21} \\ \begin{matrix} \dots \\ d_{x 1} \end{matrix} \end{matrix} & \begin{matrix} \begin{matrix} d_{12} \\ d_{22} \\ \begin{matrix} \dots \\ d_{x 2} \end{matrix} \end{matrix} & \begin{matrix} \dots \\ \dots \\ \begin{matrix} \dots \\ \dots \end{matrix} \end{matrix} \end{matrix} & \begin{matrix} d_{1 n} \\ d_{2 n} \\ \begin{matrix} \dots \\ d_{xn} \end{matrix} \end{matrix} \end{matrix}] = {(d_{ij})}_{x \times n}$ (10) $W = (W_{1}, W_{2}, \dots . . W_{n}), \sum_{i = 1}^{n} W_{i} = 1$ (11)

Where D is denoted as the production of crisp input; W is denoted as the output of the process of evaluating the high and low levels of understandability. The students provide these crisp inputs during the maximum replications with better understandability. If there is no maximum replication, then the modification process is done to improve the fuzzy optimization technique with some other values. Then the reassessment process is done to enhance the student’s understandability level, providing crisp input. The reassessment using the D is illustrated in Fig. 4.

Fig. 4

Reassessment using D.

The reassessment process is done to modify the previous values of the student’s performance and to enhance the fuzzy optimization technique to improve the understandability level of the students. The modifications are done to reassign the process to make the students understand the mathematical models and the problems. The reassessment process is done to enable the students’ understandability to a high level for providing crisp inputs by the maximum replication procedure done by the students in the performance (Fig. 4). The process of providing the maximum replication with better understandability is explained by the following Equations (12) & (13):

$K_{bj} = \frac{1}{\sum_{h = x_{j}}^{b_{j}} \frac{\sum_{i = 1}^{n} [d_{i} (V_{ij} - S_{h})]}{\sum_{i = 1}^{n} [d_{i} (V_{ij} - S_{b})]}}$ (12)

Where $V_{ij} = S_{h} (ie) {\sum_{i = 1}^{n} {[d_{i} (V_{ij} - S_{h})]}^{2}}$ . $P = \sum_{h = 1}^{a} a_{hj}^{*} . P = (P_{1}, P_{2}, \dots P_{n})$ (13)

Where K_bj is denoted as the maximum replication; S_h is denoted as the output of the process of making the crisp input; S_b is denoted as the modification and P is denoted as the reassessment process. The modification for the reassessment process is portrayed in Fig. 5.

Fig. 5

Modifications for reassessment based on k_{b
_j}.

Table 1 presents the steps of the mathematical model known as the Articulated Performance Assessment Model (APAM). The first stage is instructing students in the model and assessing their competence. The second phase entails testing for comprehension, problem-solving, and replication skills. As a third phase, we introduce fuzzy optimization, which evaluates a learner’s level of comprehension. The fourth phase involves a detailed examination of efficiency, replication, fuzzification rate, recommendation ratio, and analytical time. The sixth to eighth steps entail an extensive analysis of understandability, problem-solving, and replication skills. The last part is an optimization procedure using fuzzy logic. The pseudocode below represents the APAM model’s sequential execution.

Table 1

APAM algorithm

Algorithm 1: APAM
Step 1: Performing teaching and initial assessment
function teach_and_assess (models, students):
for each student in students:
for each model in models:
teach_student (model)
evaluate_performance (model, student)
Step: Evaluating students’ performance
function evaluate_performance (model, student):
understandability = assess_understandability (model, student)
problem_solving = assess_problem_solving (model, student)
replication = assess_replication (model, student)
Step 3: Fuzzy Optimization
fuzzy_output = fuzzy_optimization (understandability, problem_solving, replication)
Step 4: Determine students’ level of understandability
high_level = fuzzy_output.high_level
low_level = fuzzy_output.low_level
Step 5: Assessment of efficiency, replication, fuzzification rate, recommendation ratio, and analytical time
assess_efficiency (fuzzy_output)
assess_replication (fuzzy_output)
assess_fuzzification_rate (fuzzy_output)
assess_recommendation_ratio (fuzzy_output)
assess_analytical_time (fuzzy_output)
Step 6: Assessment of understandability
function assess_understandability (model, student):
# Implementation of understandability assessment
Step 7: Assessment of problem-solving skills
function assess_problem_solving (model, student):
# Implementation of problem-solving assessment
Step 8: Assessment of replication skills
function assess_replication (model, student):
# Implementation of replication assessment
Step 9: Fuzzy Optimization Process
function fuzzy_optimization (understandability, problem_solving, replication):
# Implementation of fuzzy optimization
Step 10: End.

In this method, the student’s performance is evaluated by using the fuzzy optimization technique. The analysis time is reduced and the efficiency of the students is high in the maximum replications. The fuzzification rate is high and it processes the maximum replications which produce the crisp inputs. The different factors, such as understandability, problem-solving, and replication, are agreed upon by the fuzzy optimization technique (Fig. 5). By using the fuzzy optimization technique, the understanding skills of the students can be improved. The articulated model is introduced for consenting mathematics assessment and to enhance students’ performance.

4 Discussion

The proposed model is assessed using the partnership data in [32]. This research uses partnership data to examine a model with fuzzy optimization for enhancing mathematics evaluation in higher vocational college students. The 459 pupils in the dataset range from third to eighth grade and represent diverse demographics. The data comes from 15 pedagogical modalities and 30–34 tests in each phase. The model’s flexibility is measured, and its capacities to be understood, solved, and replicated are all recorded. The data set’s primary purpose is to reveal shifts and tendencies in performance dynamics across time intervals. The data collection procedure includes 30–34 math-focused exams per grade level. The study employs the flexible and variable MDL model, emphasizing student understandability and prior learning experiences. Measures of student success in different pedagogical settings are the focus of this data collection. Therefore, the data source provides mathematics performance numerical for 15 different teaching models. These models are used independently for different grades (3 to 8) and races. This reports the performance of 459 students from different schools in two consecutive years: 2014–2015 and 2015–2016. Approximately 30–34 assessments are made focused on the grade of the students. The assessment and its model and its tenure are represented in Fig. 6.

Fig. 6

Assessment model and tenure.

The model (represented as MDL) is used for training different mathematics problems. The model varies with the student (based on understandability and previous learning) for providing abrupt session allocations. This assessment aims to unify the model for different students by consolidating a common, understandable teaching method. Based on the above model implication, the performance of the students of different grades is illustrated in Fig. 7.

Fig. 7

Performance of models and grades.

The students’ performance obtained between two consecutive years is presented in Fig. 7. The±indicates the variation compared to the previous year so that the models’ performance is handled. If depreciation is observed, the model is enlisted for 3 factors: understandability, problem-solving, and replication. These 3 factors are analyzed to identify the weak factor that is addressed using fuzzy optimization. The absolute fuzzy estimation for the 3 factors is presented in Fig. 8.

Fig. 8

Absolute estimation.

The absolute estimation is performed for F₁ = understability, F₂ = problem solving, and F₃ = replication. It is seen that replication is high regardless of different problems. This eventually increases the chances of lowering understandability. If a low fuzzy is observed, then the differentiation is performed and the maximum replicated model is identified. This is identified for selecting inputs that match the high performance of the fuzzification. Therefore, the non-absolute solution looks like the one in Fig. 9. This solution post the classification is used for identifying precise inputs.

Fig. 9

Before and after low, high classification.

The low and high classification identifies a medium point for preventing model failures. This is first identified from F₁ and its associated factors for preventing major failures. Therefore the inputs that result in achieving possible low absolutes are prevented from downfall. The inputs are precisely selected for low absolutes and are prevented from downfall. The inputs are precisely selected to maximize the efficiency of the model. Either the model is revoked or modified with suitable recommendations for performance improvements (Fig. 9). Therefore, the transition between the MDLs in the consecutive assessment session is illustrated in Fig. 10.

Based on the performance, the transition is forwarded to the next or previous MDL. If a low set of students generate low performance, then the transition is pursued using the medium absolute. Contrarily, if changes in the considered factors (i.e.) either of the factors remains stagnant, then a transition is modified with the best-known or successful MDL. Therefore, any independent model’s efficiency is retained for the next academic year. This transition break is represented by a switchover across different grades, as presented in Fig. 10.

Fig. 10

Session-based assessment model transition.

(1) Comparison Study

The comparison study is performed using the metrics efficiency, replication, fuzzification rate, analytical time, and recommendation ratio. The compared methods are AFAF [23], DIF-MMA [30], and MVCM [24].

(2) Efficiency

The efficiency of the student’s performance is better in this process by using the fuzzy optimization technique. The students’ performance is extracted from their understandability, problem-solving, and replication skills. The optimization technique is used to evaluate the students’ skills, identify their understandability level, and attain the crisp input by the maximum replications. The efficacious performance of the students can enhance their understandability skills to do the perfect replications. The teachers taught the students the mathematical models and problems to make them perform better with innovative thinking and problem-solving skills. The skills of the students can be determined from their performance after the teacher educates them about mathematical problems and problem-solving skills. Through this process, the student’s performance’s efficiency is improved and their understandability skill is improved (Fig. 11).

Fig. 11

Efficiency.

(3) Replication

The replication by the students is made high in this process by evaluating their performance after learning the mathematical problems and the models. The replications are identified by using the students’ problem-solving skill as the base. Replication is the application of problem-solving skills for mathematical models. Replication is where the different kinds and patterns of problems will be given to the students to solve and check their skills. The different iterations of the problem will be given to the student and check how they solve it by using the mathematical models or problems which the teachers educate. The varying steps in problem explanation and iteration replication are given for the students to estimate their performance. The student’s efficiency in solving the problems can be evaluated through this replication. The replications are made with the student’s high understandability level and also it provides crisp inputs at the end of the process. By this, the replications are made high by using the fuzzy optimization technique (Fig. 12).

Fig. 12

Replication.

(4) Fuzzification Rate

The fuzzification rate is high in this process by using the optimization procedure to improve the students’ performance. The fuzzification process is done to identify the students’ understandability level, whether a high level or low level. From this output, fuzzification is used to find the crisp inputs. The maximum replications can be done with the perfect crisp inputs gathered from the fuzzy optimization process outputs. The optimization recommends the change in different steps in problem explanation and iterated replication to identify the students’ performance. This process fluctuates between irrelevant and crisp inputs during fuzzification. In this process, the crisp inputs are the maximum replications produced by the students with better understanding skills. The fuzzification is used to improve the innovative skills of the students for solving the problems by performing a better replications process during the evaluation procedure. By this process, the fuzzification rate is made better in this evaluation of the student’s performance method (Fig. 13).

Fig. 13

Fuzzification rate.

(5) Analytical Time

The time taken for the process of analyzing is less in this process by using the optimization process. The evaluation time of the student’s performance is taken less with the help of the optimization process. After the teachers educated the mathematical models and the problems to the students, the assessment is conducted to evaluate the performance of the students. From their performance the understandability, the problem solving and replication skills of the students are extracted by the fuzzy optimization technique. Then the students’ level of things can be found by using the replication process. The replication is the one which is a different kind of problem given to the students to determine their innovative thinking. Through that process, the efficiency of the students in solving the problems can be identified and if there is any lag in the understandability, then further steps are taken to enhance their problem-solving skills and their performance. Then the maximum replications are done to enhance the crisp inputs with the better understanding skills of the students (Fig. 14).

Fig. 14

Analytical time.

(6) Recommendation Ratio

The recommendation ratio is high in this process by using the fuzzy optimization technique in the students’ performance evaluation process. APAM for consenting mathematics assessment is introduced for subscribing to the mathematics assessments. Fuzzy optimization is used for agreeing on different factors such as understandability, problem-solving, and replication. The students’ understandability is determined using similar problem progression, whereas replication is the application of problem-solving skills for coherent mathematical models. From the outlook, problem-solving, and decision-making, the students must study the idea. The assessment, which is the output of the student’s performance due to the fuzzy optimization, is evaluated for high and low-level understandability. The optimization recommends the change in different steps in problem explanation and iterated replication to leverage the students’ performance. By this, the recommendation ratio is high in this process by using the optimization technique (Fig. 15).

Fig. 15

Recommendation ratio.

5 Conclusion

This article introduced an articulated model using fuzzy optimization to improve the quality of higher vocational college students’ mathematics performance assessment. This model relies on the students’ understandability, problem-solving, and replication metrics. The fuzzy model performs an iterated evaluation for varying steps in problem-solving and understandability checks. The understandability for low and high levels is segregated through maximum model replications with precise modifications and assessments. In the modification, the recommendations are implied for filtering crisp inputs. This prevents overpopulated and complex computations due to different models and patterns exhibited. Considering the iteration across the varying replication, the proposed model improves the performance through modified pattern differentiation. Therefore, the proposed model excludes the modification-causing inputs in successive iterations to prevent model failures. This proposed model improves efficiency, replication, fuzzification rate, and recommendation ratio by 12.11%, 12%, 9.01%, and 8.66%, respectively. It reduces the analytical time by 10.41% under the varying models.

References

O’Sullivan

, Casey

and Crowley

, Asynchronous online mathematics learning support: an exploration of interaction data to inform future provision, Teaching Mathematics and its Applications: An International Journal of the IMA 40(4) (2021), 317–331.

Sabir

, Saoud

, Raja

M.A.Z.

, Wahab

H.A.

and Arbi

, Heuristic computing technique for numerical solutions of nonlinear fourth order Emden-Fowler equation, Mathematics and Computers in Simulation 178 (2020), 534–548.

Kahl

, Segerer

, Grob

and Möhring

, Bidirectional associations among executive functions, visual-spatial skills, and mathematical achievement in primary school students: Insights from a longitudinal study, Cognitive Development 62 (2022), 101149.

Sellers

M.E.

, Roh

K.H.

and Parr

E.D.

, Student quantifications as meanings for quantified variables in complex mathematical statements, The Journal of Mathematical Behavior 61 (2021), 100802.

Haser

Ç.

, Doğan

and Erhan

G.K.

, Tracing students’mathematics learning loss during school closures in teachers’self-reported practices, International Journal of Educational Development 88 (2022), 102536.

Dunn

P.K.

and Marshman

M.F.

, Teaching mathematical modelling: a framework to support teachers’ choice of resources, Teaching Mathematics and its Applications: An International Journal of the IMA 39(2) (2020), 127–144.

Simsek

, Jones

, Hunter

and Xenidou-Dervou

, Mathematical equivalence assessment: Measurement invariance across six countries, Studies in Educational Evaluation 70 (2021), 101046.

Ehmke

, van den Ham

A.K.

, Sälzer

, Heine

and Prenzel

, Measuring mathematics competence in international and national large scale assessments: Linking PISA and the national educational panel study in Germany, Studies in Educational Evaluation 65 (2020), 100847.

Tam

Y.P.

and Chan

W.W.L.

, The differential relations between sub-domains of spatial abilities and mathematical performance in children, Contemporary Educational Psychology 71 (2022), 102101.

10.

Cai

and Yang

, The fluid relation between reading strategies and mathematics learning: A perspective of the Island Ridge Curve, Learning and Individual Differences 98 (2022), 102180.

11.

Grove

, Croft

and Lawson

, The extent and uptake of mathematics support in higher education: results from the survey, Teaching Mathematics and its Applications: An International Journal of the IMA 39(2) (2020), 86–104.

12.

Attard

and Holmes

, “It gives you that sense of hope”: An exploration of technology use to mediate student engagement with mathematics, Heliyon 6(1) (2020), e02945.

13.

Y.Y.

and Rose

M.K.

, Considering Proofs: Pre-Service Secondary Mathematics Teachers’ Criteria for Self-Constructed and Student-Generated Arguments, The Journal of Mathematical Behavior 68 (2022), 100999.

14.

Hwang

G.J.

, Wang

S.Y.

and Lai

C.L.

, Effects of a social regulation-based online learning framework on students’ learning achievements and behaviors in mathematics, Computers and Education 160 (2021), 104031.

15.

Rolfe

, Hansen

K.Y.

and Strietholt

, Integrating educational quality and educational equality into a model of mathematics performance, Studies in Educational Evaluation 74 (2022), 101171.

16.

Omaish

H.A.

, Hosein

, Abdullah

M.U.

and Aldershewi

, University lecturer’s perceptions on the causes of students’ mathematical knowledge gaps in conflict zones, International Journal of Educational Research Open 2 (2021), 100095.

17.

S.H.

and Jitendra

A.K.

, Analysis of proportional reasoning and misconceptions among students with mathematical learning disabilities, The Journal of Mathematical Behavior 57 (2020), 100753.

18.

Legesse

, Luneta

and Ejigu

, Analyzing the effects of mathematical discourse-based instruction on eleventh-grade students’ procedural and conceptual understanding of probability and statistics, Studies in Educational Evaluation 67 (2020), 100918.

19.

Huey

M.E.

, Assessing the impact of standards-based grading policy changes on student performance and practice work completion in secondary mathematics, Studies in Educational Evaluation 75 (2022), 101211.

20.

Song

C.S.

, Xu

, Maloney

E.A.

, Skwarchuk

S.L.

, Burr

S.D.L.

, Lafay

and LeFevre

J.A.

, Longitudinal relations between young students’ feelings about mathematics and arithmetic performance, Cognitive Development 59 (2021), 101078.

21.

Hidayat

, Zamri

S.N.A.S.

, Zulnaidi

and Yuanita

, Meta-cognitive behaviour and mathematical modelling competency: Mediating effect of performance goals, Heliyon 6(4) (2020), e03800.

22.

Wang

, Cho

H.J.

, Wiles

, Moss

J.D.

, Bonem

E.M.

, Li

and Levesque-Bristol

, Competence and autonomous motivation as motivational predictors of college students’ mathematics achievement: from the perspective of self-determination theory, International Journal of STEM Education 9(1) (2022), 1–14.

23.

Moreno

and Pineda

A.F.

, A framework for automated formative assessment in mathematics courses, IEEE Access 8 (2020), 30152–30159.

24.

Nguyen-Huy

, Deo

R.C.

, Khan

, Devi

, Adeyinka

A.A.

, Apan

A.A.

and Yaseen

Z.M.

, Student performance predictions for advanced engineering mathematics course with new multivariate copula models, IEEE Access 10 (2022), 45112–45136.

25.

Wijaya

T.T.

, Cao

, Weinhandl

, Tamur

, A Meta-Analysis of the Effects of E-Books on students’ mathematics achievement, Heliyon 2022, e09432.

26.

Ding

and Homer

, Interpreting mathematics performance in PISA: Taking account of reading performance, International Journal of Educational Research 102 (2020), 101566.

27.

Sun

, Ruokamo

, Siklander

, Li

and Devlin

, Primary school students’ perceptions of scaffolding in digital game-based learning in mathematics, Learning, Culture and Social Interaction 28 (2021), 100457.

28.

Lahdenperä

, Rämö

and Postareff

, Student-centred learning environments supporting undergraduate mathematics students to apply regulated learning: A mixed-methods approach, The Journal of Mathematical Behavior 66 (2022), 100949.

29.

Sabir

, Raja

M.A.Z.

, Alhazmi

S.E.

, Gupta

, Arbi

and Baba

I.A.

, Applications of artificial neural network to solve the nonlinear COVID-19 mathematical model based on the dynamics of SIQ, Journal of Taibah University for Science 16(1) (2022), 874–884.

30.

Cascella

, Giberti

and Viale

, Investigating foreign students’ disadvantage in mathematics: A mixed method analysis to identify features of items favouring native students, The Journal of Mathematical Behavior 67 (2022), 100990.

31.

Sabir

, Hashem

A.F.

, Arbi

and Abdelhalim

M.A.

, Designing a Bayesian regularization approach to solve the fractional layla and Majnun system, Mathematics 11(17) (2023), 3792.

32.

https://data.world/mkalish/partnership-for-assessment-ofreadiness-for-college-and-careers-parcc